Packet-Type Aware Scheduling of Moldable Streaming Tasks on Multicore Systems with DVFS
Boulasikis, Michail LU ; Kessler, Christoph ; Gruian, Flavius LU
; Keller, Jorg W.
and Litzinger, Sebastian
(2024)
39th Annual ACM Symposium on Applied Computing, SAC 2024
p.449-451
- Abstract
We introduce a novel approach for energy-efficient scheduling of data-dependent stream programs with packet types on multicore architectures with voltage and frequency scaling. To have the given application meet specific throughput demands while minimizing energy consumption, we enhance the existing crown scheduling by packet-type dependent parameters. Formulation as an integer linear program generates a parametric, multi-scenario crown schedule for table-driven execution. By inspecting the packet stream at runtime, our online scheduler selects the appropriate schedule and handles frequency levels. The performance of our technique is compared to crown scheduling with synthetic stream programs. Our results demonstrate an average energy... (More)
We introduce a novel approach for energy-efficient scheduling of data-dependent stream programs with packet types on multicore architectures with voltage and frequency scaling. To have the given application meet specific throughput demands while minimizing energy consumption, we enhance the existing crown scheduling by packet-type dependent parameters. Formulation as an integer linear program generates a parametric, multi-scenario crown schedule for table-driven execution. By inspecting the packet stream at runtime, our online scheduler selects the appropriate schedule and handles frequency levels. The performance of our technique is compared to crown scheduling with synthetic stream programs. Our results demonstrate an average energy reduction of 23.94%.
(Less)
- author
- Boulasikis, Michail
LU
; Kessler, Christoph
; Gruian, Flavius
LU
; Keller, Jorg W.
and Litzinger, Sebastian
- organization
- publishing date
- 2024-04
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing
- pages
- 449 - 451
- publisher
- Association for Computing Machinery (ACM)
- conference name
- 39th Annual ACM Symposium on Applied Computing, SAC 2024
- conference location
- Avila, Spain
- conference dates
- 2024-04-08 - 2024-04-12
- external identifiers
-
- scopus:85197703464
- ISBN
- 979-840070243-3
- DOI
- 10.1145/3605098.3636081
- language
- English
- LU publication?
- yes
- id
- 0e55a6ce-b141-4877-8db5-91120b8c8e78
- date added to LUP
- 2024-10-28 12:30:16
- date last changed
- 2025-04-04 14:17:34
@inproceedings{0e55a6ce-b141-4877-8db5-91120b8c8e78,
abstract = {{<p>We introduce a novel approach for energy-efficient scheduling of data-dependent stream programs with packet types on multicore architectures with voltage and frequency scaling. To have the given application meet specific throughput demands while minimizing energy consumption, we enhance the existing crown scheduling by packet-type dependent parameters. Formulation as an integer linear program generates a parametric, multi-scenario crown schedule for table-driven execution. By inspecting the packet stream at runtime, our online scheduler selects the appropriate schedule and handles frequency levels. The performance of our technique is compared to crown scheduling with synthetic stream programs. Our results demonstrate an average energy reduction of 23.94%.</p>}},
author = {{Boulasikis, Michail and Kessler, Christoph and Gruian, Flavius and Keller, Jorg W. and Litzinger, Sebastian}},
booktitle = {{Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing}},
isbn = {{979-840070243-3}},
language = {{eng}},
pages = {{449--451}},
publisher = {{Association for Computing Machinery (ACM)}},
title = {{Packet-Type Aware Scheduling of Moldable Streaming Tasks on Multicore Systems with DVFS}},
url = {{http://dx.doi.org/10.1145/3605098.3636081}},
doi = {{10.1145/3605098.3636081}},
year = {{2024}},
}